AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Sketch-BERT: Learning Sketch Bidirectional Encoder Representation From Transformers by Self-Supervised Learning of Sketch Gestalt

Hangyu Lin, Yanwei Fu, Xiangyang Xue, Yu-Gang Jiang

CVPR 2020 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 59 citations (OpenAlex)

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add sketch Gestalt self-supervised pre-training (ablation)Sketch-BERT · QuickDraw classificationTop-1 accuracy: 83.10 -> 88.30Improved
add sketch Gestalt self-supervised pre-training (ablation)Sketch-BERT · TU-Berlin classificationTop-1 accuracy: 54.20 -> 76.30Improved
add sketch Gestalt self-supervised pre-training (ablation)Sketch-BERT · QuickDraw retrievalmAP: 55.10 -> 78.87Improved
add sketch Gestalt self-supervised pre-training (ablation)Sketch-BERT · TU-Berlin retrievalmAP: 24.14 -> 41.54Improved
use full position-and-state sketch Gestalt masking rather than BERT-style random masking (ablation)Sketch-BERT pre-training · QuickDraw classificationTop-1 accuracy: 86.51 -> 88.30Improved
mask both positions and states rather than positions only (ablation)Sketch-BERT pre-training · QuickDraw classificationTop-1 accuracy: 87.37 -> 88.30Improved
mask positions and states rather than states only (ablation)Sketch-BERT pre-training · QuickDraw retrievalTop-1 accuracy: 81.87 -> 85.47Improved
increase pre-training samples per class from 5K to 70K (ablation)Sketch-BERT · QuickDraw classificationTop-1 accuracy: 85.73 -> 88.30Improved
increase pre-training categories from 100 to 200 at 5K samples per class (ablation)Sketch-BERT · QuickDraw classificationTop-1 accuracy: 85.82 -> 84.89Hurt
increase transformer depth from 6 to 12 layers at hidden size 256 (ablation)Sketch-BERT · QuickDraw classificationTop-1 accuracy: 84.83 -> 86.34Improved
use 12-16-1024 instead of 8-12-768 architecture (ablation)Sketch-BERT · QuickDraw classificationTop-1 accuracy: 88.30 -> 85.31Hurt
use Sketch-BERT rather than TCNetsketch retrieval · QuickDrawmAP: 76.38 -> 78.87Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy